Volume 80, Issue 9, Pages 926-937 (November 2011) Slc26a11, a chloride transporter, localizes with the vacuolar H+-ATPase of A- intercalated cells of the kidney  Jie Xu, Sharon Barone, Hong Li, Shannon Holiday, Kamyar Zahedi, Manoocher Soleimani  Kidney International  Volume 80, Issue 9, Pages 926-937 (November 2011) DOI: 10.1038/ki.2011.196 Copyright © 2011 International Society of Nephrology Terms and Conditions

Figure 1 Slc26a11 distribution in mouse tissues and kidney. (a) Tissue distribution of Slc26a11. Slc26a11 transcript size was 2.0kb. Slc26a11 was detected abundantly in the kidney and brain, followed by upper small intestine and distal colon. 28S rRNA levels are shown as constitutive controls. In all, 30μg of RNA was loaded on each lane. (b) Kidney expression of Slc26a11 by RT-PCR. A representative ethidium bromide staining of agarose gel demonstrates a PCR product of expected size (1200bp) in the cortex and medulla. (c) Tissue distribution of the Slc26a11 long variant. The long Slc26a11 variant is expressed in the kidney and other epithelial cells, with lower levels in the brain. (d) Tissue distribution of the Slc26a11 short variant. The short Slc26a11 variant is predominantly expressed in the brain, with lower levels in other tissues. RT-PCR, reverse transcriptase-PCR. Kidney International 2011 80, 926-937DOI: (10.1038/ki.2011.196) Copyright © 2011 International Society of Nephrology Terms and Conditions

Figure 2 Immunolocalization of Slc26a11 (KBAT) in the kidney. (a) Specificity of KBAT antibodies and immunoblotting of KBAT in the kidney. Left panel: with the use of the antibody generated against the mouse Slc26a11, a ∼72-kDa band was detected in the kidney medulla. Labeling of the 72-kDa band was completely prevented with the pre-adsorbed immune serum. Right panel: KBAT antibody specifically labels the plasma membrane of COS7 cells transfected with Slc26a11 cDNA. Top: cells transfected with empty vector (sham), bottom: cells transfected with Slc26a11 cDNA. Cells were labeled with KBAT antibodies (Slc26a11, left) and co-labeled with phalloidin (right), as a marker of the actin cytoskeleton. (b) Immunofluorescent labeling with KBAT antibodies (low magnification). Low magnification of Slc26a11 labeling in the cortex (A, left), outer medulla (B, middle), and the inner medulla (C, right). (c) Immunofluorescent labeling with KBAT antibodies (high magnification). Higher magnifications of Slc26a11 labeling in the cortex (A, left), outer medulla (B, middle), and the inner medulla (C, right). Results clearly demonstrate that in the cortex, the labeling of KBAT was detected on the apical or basolateral membrane domains in the CCD (arrows). In the medulla, KBAT was only detected on the apical membrane domain of a subset of cells in OMCD (B) and IMCD (C) (arrows). (d) Immunolabeling with pre-adsorbed immune serum. No labeling was detected with the pre-adsorbed immune serum in the cortex (A) and in the medulla (B and C). CCD, cortical collecting duct; IMCD, inner medullary collecting duct; KBAT, kidney brain anion transporter; OMCD, outer medullary collecting duct. Kidney International 2011 80, 926-937DOI: (10.1038/ki.2011.196) Copyright © 2011 International Society of Nephrology Terms and Conditions

Figure 3 Immunofluorescent double staining of KBAT with AQP2 or H+-ATPase. (a) Double immunofluorescent labeling of KBAT and aquaporin 2 (AQP2) in the kidney. (A) Cortex (top). Immunostaining in the kidney indicated that the distribution of AQP2 (right, purple arrow) and Slc26a11 (left, white arrow) corresponded to two distinct cell subtypes in the CCD when dual images were acquired (middle). (B and C) Outer medulla (B) and inner medulla (C). Immunostaining in the kidney indicated that the distribution of AQP2 (right, purple arrows) and Slc26a11 (left, white arrows) corresponded to two distinct cell subtypes in the OMCD (B) and IMCD (C) when dual images were acquired (middle), indicating that Slc26a11 is localized on the apical membrane of cells distinct from principal cells. (b) Double-immunofluorescent labeling of KBAT and H+-ATPase in the kidney. (A) Cortex. Two sets of data (top and bottom panels) are shown in panel (bA). As shown, Slc26a11 (left, white arrows) and H+-ATPase (right, purple arrows) demonstrate remarkably identical localization patterns in the cortical collecting duct when merged images are acquired (middle). Slc26a11 co-localizes with H+-ATPase on the apical membrane of A-intercalated cells and on the basolateral membrane of B-intercalated cells in the CCD. (B and C) Outer medulla (bB) and inner medulla (bC). In the OMCD (bB) and IMCD (bC), Slc26a11 (left, white arrows) co-localizes with H+-ATPase (right, purple arrows) exclusively on the apical membrane of A-intercalated cells (merged image in the middle). CCD, cortical collecting duct; IMCD, inner medullary collecting duct; KBAT, kidney brain anion transporter; OMCD, outer medullary collecting duct. Kidney International 2011 80, 926-937DOI: (10.1038/ki.2011.196) Copyright © 2011 International Society of Nephrology Terms and Conditions

Figure 4 Functional identity of Slc26a11 (KBAT). COS7 cells were transiently transfected with KBAT cDNA and studied by 36Cl flux assay (a–c) and intracellular pH monitoring (d and e). (Panel a) 36Cl influx. Cells were transiently transfected with Slc26a11 (KBAT) cDNA and assayed 48h later. The influx of 2mmol/l 36Cl was stopped at 10min using cold saline. The nonspecific Cl uptake (background) was calculated in both transfected and non-transfected cells at time 0 and deducted from their respective 10-min influx values. As shown, cells transfected with KBAT demonstrated significant 36Cl influx relative to mock-transfected cells. The presence of DIDS significantly inhibited the 36Cl influx. Extracellular K=4mEq. (Panel b) 36Cl efflux (in the absence or the presence of external chloride). Cells were transiently transfected with Slc26a11 (KBAT) cDNA and assayed 48h later. Cells were first pre-incubated with 2mmol/l 36Cl for 30min. Thereafter, the radioactive media were aspirated and cells were washed three times in rapid succession and then incubated with a chloride-free or a chloride-containing solution. The reaction was stopped at time 0 or 10min. The chloride efflux was calculated as the difference in cell chloride count at time zero (control) and 10min. As shown, cells transfected with KBAT demonstrated significant amount of 36Cl efflux relative to mock-transfected cells, when an outwardly directed potassium gradient (extracellular K=1mEq) was imposed. In the presence of external chloride (100mmol/l), the rate of 36Cl efflux was significantly increased relative to no external chloride. (Panel c) Effect of voltage clamping on 36Cl efflux in the presence or absence of external chloride. Cells were transiently transfected and then loaded with 2mmol/l 36Cl for 30min as above. Cells were pre-incubated with 10μmol/l valinomycin or ethanol (as vehicle). The 10-min efflux of 36Cl was assayed in the presence or absence of potassium-rich (100 or 5mmol/l KCl) solution. The unidirectional chloride efflux by KBAT was abrogated, and the anion exchange mode of KBAT was partially inhibited under voltage-clamped conditions. (Panel d) Intracellular pHi monitoring. Cl−/HCO3− exchanger activity was assayed by subsequent removal and addition of Cl− as described in the ‘Materials and Methods’ section. The experiments were repeated in the absence of CO2/HCO3− (see below). Both the intracellular alkalinization upon chloride removal and the pHi recovery upon switching back to chloride-containing solution were increased in KBAT-transfected cells. In the absence of bicarbonate, the rate of pHi alteration was minimal in both sham and KBAT-transfected cells (below). (Panel e) Summation of pHi tracing studies. In the presence of CO2/HCO3− in the media, the rate of chloride-dependent bicarbonate transport was significantly increased in KBAT-transfected cells relative to mock-transfected cells. In the absence of CO2/HCO3− in the media, the rate of chloride-dependent pHi alteration was minimal in both KBAT-transfected and non-transfected cells. Buffering capacity (βi) was calculated in mock and KBAT-transfected cells according to established protocols and was not different between the two groups. DIDS, 4,4′-Diisothiocyano-2,2′-stilbenedisulfonic acid; KBAT, kidney brain anion transporter; pHi, intracellular pH; Valino, valinomycin. Kidney International 2011 80, 926-937DOI: (10.1038/ki.2011.196) Copyright © 2011 International Society of Nephrology Terms and Conditions

Figure 5 Effect of Slc26a11 (KBAT) on H+-ATPase activity (pH studies). (a) Intracellular pH tracing. Cultured COS7 cells were acid loaded with NH4+ pulse in the presence of Na-free (TMA)-Cl. Switching from the isotonic, Na-free solution to hypotonic, Na-free solution resulted in pHi recovery in mock-transfected, acid-loaded cells (left). KBAT-transfected cells showed a more robust recovery from the acid-loaded state when switched to the hypotonic Na-free solution (right). (b) KBAT-transfected vs mock-transfected cells: Summation of results. Summation of recovery rates from acidosis from six separate coverslips showed that H+-ATPase-mediated pHi recovery from intracellular acidosis was significantly increased in KBAT-transfected cells. (c) Characterization of KBAT activation of H+-ATPase. Effect of chloride-free solution on KBAT activated H+-ATPase. Control and KBAT-transfected COS7 cells were acid loaded, depleted of chloride, and then monitored for the recovery from intracellular acidosis in the absence of chloride in the external solution. KBAT stimulation of H+-ATPase activity was almost abrogated in the chloride-depleted state. Effect of DIDS on KBAT activated H+-ATPase. pHi recovery from intracellular acidosis was monitored in control and KBAT-transfected COS7 cells in the presence of 0.5mmol/l DIDS added to the external solution. Effect of voltage clamping on KBAT activated H+-ATPase. Control and KBAT-transfected COS7 cells were pre-incubated with valinomycin at 10μmol/l. Cells were acid loaded and then monitored for recovery from intracellular acidosis in a high-potassium, chloride-containing solution. As shown, KBAT stimulation of H+-ATPase was partially inhibited under voltage clamped conditions. DIDS, 4,4′-Diisothiocyano-2,2′-stilbenedisulfonic acid; KBAT, kidney brain anion transporter; pHi, intracellular pH. Kidney International 2011 80, 926-937DOI: (10.1038/ki.2011.196) Copyright © 2011 International Society of Nephrology Terms and Conditions